Retinoic acid (RA) is a cell-cell signaling molecule derived from retinol that controls several aspects of development. As RA is useful for differentiation of embryonic stem cells, studies designed to understand how RA performs its functions will be essential for development of stem cell-based treatments for disease. In order to generate replacement organs from stem cells we need to understand how embryos initially generate organs, including knowledge of the regulatory molecules that send signals between cells. RA signaling occurs when retinol is sequentially metabolized to retinaldehyde and then to RA which functions as a ligand for nuclear RA receptors that bind DNA and directly regulate gene expression. The regulatory enzyme controlling synthesis of RA during early embryogenesis has been discovered to be retinaldehyde dehydrogenase-2 (RALDH2) which oxidizes retinaldehyde to RA, initially only in posterior mesoderm. RALDH2 is conserved in human, mouse, chick, frog, and fish, and is the only enzyme synthesizing RA in early mouse embryos. Raldh2 null mutant mouse embryos lose all RA signaling activity normally present in the posterior portion of the embryo from the late primitive streak stage to the tailbud stage and do not develop beyond the early tailbud stage. These mutants exhibit disrupted somite segmentation and fail to generate forelimb buds, indicating that RA is required for proper development of these mesodermal tissues. Low-dose maternal dietary RA supplementation can rescue these defects. Our laboratory has discovered that a quite useful tool for unraveling RA function is to provide RA supplementation to Raldh2 mutants for various lengths of time and then follow where the exogenous RA stimulates transcription using an RA- reporter transgene also present in these embryos. Those studies have allowed us to hypothesize that Raldh2 functions only in a cell-nonautonomous fashion, meaning that RA synthesized in the mesoderm is secreted and acts on neighboring cells, but does not function within the cells producing RA. This important piece of information must now be integrated into our view of how RA controls mesodermal development during early development. In this proposal, Raldh2-/- mice carrying an RA-reporter transgene will be used as a model system for revealing the spatiotemporal mechanism of RA action during mesoderm development and the genes regulated. Raldh2-/- mice will be crossed with various other null mutant mice to address the mechanism of RA action. The specific aims are as follows: (1) we will determine the mechanism that directs RA action to the neuroectoderm and prevents RA activity in the presomitic mesoderm of RA-rescued Raldh2-l- embryos. (2) We will examine the mechanism of RA action required for proper somite formation and patterning including an analysis of the genes regulated by RA in the adjacent neuroectoderm and the signals sent from neuroectoderm to presomitic mesoderm. (3) The role RA signaling plays in forelimb budding will be investigated including the timing, target tissue, RA source tissue, and target genes.